Educate, Decide, Use

*Originally published in the July edition of the Voluntary Benefits Voice.*

By Ben Yomtoob and Meg Collins

The Educate / Decide / Use framework (EDU) organizes the benefits guidance challenge into three constituent events. When we introduced it last month, we argued that most tools cover one moment well and leave the others to chance. Readers agreed, then asked the obvious follow-up: What is the current state of play for each? The answer is that the three moments are unevenly served. Use is the one that matters most once coverage is in force, and yet Use is the furthest behind.

Educate: The Moment Everyone Assumes Is Handled

Educate is the easiest event to claim and the hardest to verify. Nearly every benefits communication vendor, administration platform, and carrier portal includes some version of an education layer: plan summaries, short videos, or a chatbot trained on a benefits glossary. The bar to entry is low: producing content that explains deductibles isn't something that requires deep integration with plan documents or claims systems.

That low bar is part of the problem.

Comprehension content built once at the start of a plan year, and not recontextualized as workers' needs evolve throughout the year, is less than adequate. LIMRA finds that 73% of employees want to hear about their benefits more often, yet half hear about them only at open enrollment [1]. Educate-moment guidance must be persistent and contextual: available the day an employee gets married, the day a new dependent is added, the week before a high-deductible plan's coinsurance resets.

As an example of the shortcomings of today's tools, LIMRA finds that supplemental health and disability plans are the least understood benefits offering, and that employees who don't understand a benefit may assume something is covered when it isn't, or may neglect to file a claim they were eligible for [1]. That is not a Decide-moment failure; these employees were enrolled. Nor is it a Use-moment failure in the mechanical sense: they never reach for the claim form, because they never understood what the voluntary policy was meant to cover. It is an Educate-moment failure, persisting year-round, on a product the worker is already paying for.

The deeper failure is that Educate rarely operates as a discrete event at all: it collapses into Decide. The two get lumped together at 4:30pm on the last day of open enrollment, when the employee finally opens the portal, and tries to learn the plan and choose it in the same rushed session. Education becomes a formality the worker rushes through on the way to clicking "enroll".

Decide: The Most Crowded, Most Funded, Most Visible Moment

Decide is where the market has concentrated its capital and its attention, for a straightforward reason: it is the moment with the clearest commercial hook.

A tool that helps an employee choose a plan can point to enrollment completion rates, time-to-enroll, and adoption of voluntary products as concrete, demonstrable outcomes. Carriers and brokers can see the effect in the numbers they already track. Cost calculators, side-by-side comparison tools and recommendation engines have all built their value proposition around this single event, and the result is a competitive, fast-iterating category.

This category is also where AI adoption is most visible. Alight reported 8.5 million AI chatbot interactions during 2025 open enrollment, roughly three times the prior year, and found that 43% of workers now trust AI for benefits advice [2]. Businessolver reported that 84% of employees rated their enrollment experience "great" or "excellent," with the ones using AI guidance 129% more likely to enroll in an HSA [3].

Unfortunately, these are activity and experience metrics rather than outcome metrics: they describe what employees did and how they felt about it, not whether they ended up in the right plan, or whether their enrollment choices improved utilization.

Use: The Most Underserved Moment and the Hardest to Prove

Use is where the guidance stack falls apart, and it is also where the consequences of that failure are most concrete. This is the event that determines whether the coverage an employee selected actually functions when it comes time to file a claim. For voluntary benefits specifically, Use is the event tied most directly to persistency and renewal, because an employee who never used a policy is an employee who is unlikely to renew it.

It is also the event with the thinnest tooling. Most benefits administration platforms end their guidance functionality at the enrollment confirmation screen. Care navigation vendors cover medical claims reasonably well, but rarely extend the same depth to voluntary products.

The downstream cost of the Use gap is visible in the data, even without claims utilization numbers. 32.1% of members report having no primary care provider at all, up nearly two points since 2024 [4]. This is a direct measure of disengagement from the preventive side of the Use moment, since deferred care simply becomes more expensive acute care later. This dynamic compounds for supplemental coverage: disengaged members skip preventive care, overuse the emergency room, and do not file the supplemental health claims for the voluntary products they are already enrolled in.

The more significant problem is that the data needed to measure the Use moment is largely unavailable. Enrollment data is collected obsessively, reported in every benchmarking study, and cited in every vendor pitch deck.

Claims utilization data -- who files a claim after enrolling in a voluntary product, and whether they even understood how to do so -- is not publicly reported in any comparable way. This point shows up in the very questions employers and brokers are asking right now: whether employees ended up in the right plan and whether utilization actually improved. The real answer across the market is: nobody currently knows.

What the Framework Exposes

Laid side by side, the three EDU events tell a consistent story. Educate is widely claimed and weakly evidenced. Decide is heavily resourced and validated against an incomplete endpoint. Use is the most consequential for voluntary benefits and the most thinly covered.

None of this means existing tools are without value. It means that the language used by vendors ("AI-powered guidance," "decision support," "benefits assistant") describes very little about which of the three events a tool is actually built to serve, and even less about how well it does it. A buyer evaluating a vendor's claim to cover the full employee journey should ask three direct questions for each EDU event:

What does this tool do here? What evidence supports that it works? And, is that evidence independently verifiable or only self-reported?

Those questions are the starting point for the next phase of our work. The Benefits Guidance Consortium's forthcoming State of the Market white paper will apply this lens across the vendor landscape, organizing tools by where they actually operate, rather than by the category name they have chosen for themselves. Future installments in this series will look more closely at the disclosure problem in Decide-moment recommendation engines, and at what it would take to build the data infrastructure the Use moment is currently missing.

---

*About the Authors: Ben Yomtoob and Meg Collins co-founded the Benefits Guidance Consortium to define and promote the Benefits Guidance category. Ben is Founder / Lead Consultant at BuckleyRoberts and a former CEO of Workterra. Meg is Founder and CEO of RevGem Consulting and a former Chief Growth Officer at Ideon.*

Sources

[1] LIMRA | 2024 BEAT (Benefits and Employee Attitude Tracker) Study -- limra.com

[2] Alight | 2025 Annual Enrollment Analysis -- alight.com

[3] Businessolver | 2024 Post Enrollment Survey -- businessolver.com

[4] 32.1% of members report having no primary care provider -- source cited in original publication